1 Rethinking Data Management for Storage-centric Sensor Networks Yanlei Diao, Deepak Ganesan, Gaurav Mathur, and Prashant Shenoy CIDR 2007 Proceedings.

Slides:



Advertisements
Similar presentations
Distributed Data Processing
Advertisements

anywhere and everywhere. omnipresent A sensor network is an infrastructure comprised of sensing (measuring), computing, and communication elements.
System Design Issues In Sensor Databases Qiong Luo and Hejun Wu Department of Computer Science and Engineering The Hong Kong University of Science & Technology.
Kien A. Hua Division of Computer Science University of Central Florida.
Wireless Sensor Networks: An overview and experiences. Matthew Grove PEDAL Seminar Series, January 9th 2008.
Semantic Web Based Architecture for Managing Hardware Heterogeneity in Wireless Sensor Network Authors: Sinisa Nikolić, MSc Valentin Penca, MSc Milan Segedinac,
An Evaluation of Multi-Resolution Storage for Sensor Networks SenSys’03 Paper by Deepak Ganesan, Ben Greenstein, Denis Perelyubskiy, Deborah Estrin, and.
Panoptes: A Scalable Architecture for Video Sensor Networking Applications Wu-chi Feng, Brian Code, Ed Kaiser, Mike Shea, Wu-chang Feng (OGI: The Oregon.
University of Massachusetts, Amherst Triage: Balancing Energy and Quality of Service in a Microserver Nilanjan Banerjee, Jacob Sorber, Mark Corner, Sami.
Course Project Ideas Yanlei Diao University of Massachusetts Amherst.
Cougar (Mica Mote) A platform for testing query processing techniques over ad-hoc sensor networks Three tier system: – Running TinyOS, an embedded operating.
1 Storing Data: Disks and Files Yanlei Diao UMass Amherst Feb 15, 2007 Slides Courtesy of R. Ramakrishnan and J. Gehrke.
Last Time –Main memory indexing (T trees) and a real system. –Optimize for CPU, space, and logging. But things have changed drastically! Hardware trend:
1 An Evaluation of Multi-resolution Storage for Sensor Networks Deepak Ganesan, Ben Greenstein, Denis Perelyubskiy, Deborah Estrin (UCLA), John Heidemann.
Generic Sensor Platform for Networked Sensors Haywood Ho.
UNIVERSITY OF SOUTHERN CALIFORNIA Embedded Networks Laboratory 1 Wireless Sensor Networks Ramesh Govindan Lab Home Page:
1 An Evaluation of Multi-resolution Storage for Sensor Networks D. Ganesan, B. Greenstein, D. Perelyubskiy, D. Estrin, J. Heidemann ACM SenSys 2003.
4/30/031 Wireless Sensor Networks for Habitat Monitoring CS843 Gangalam Vinaya Bhaskar Rao.
Query and Storage in Wireless Sensor Networks. The Problem ◊How to perform efficient query and storage in wireless sensor networks? ◊Design goals:  Distributed.
U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Re-thinking Data Management for Storage-Centric Sensor Networks Deepak Ganesan University.
Computer Science Storage Systems and Sensor Storage Research Overview.
Energy-efficient Self-adapting Online Linear Forecasting for Wireless Sensor Network Applications Jai-Jin Lim and Kang G. Shin Real-Time Computing Laboratory,
Department of Computer Science University of Massachusetts, Amherst PRESTO: Feedback-driven Data Management in Sensor Network Ming Li, Deepak Ganesan,
Gordon: Using Flash Memory to Build Fast, Power-efficient Clusters for Data-intensive Applications A. Caulfield, L. Grupp, S. Swanson, UCSD, ASPLOS’09.
Moving Objects Databases Nilanshu Dharma Shalva Singh.
The Platforms enabling Wireless Sensor Networks Hill, Horton, Kling, Krishnamurthy CACM, June 2004.
Sensor Networks Storage Sanket Totala Sudarshan Jagannathan.
SensEye: A Multi-Tier Camera Sensor Network by Purushottam Kulkarni, Deepak Ganesan, Prashant Shenoy, and Qifeng Lu Presenters: Yen-Chia Chen and Ivan.
TSAR: A Two Tier Sensor Storage Architecture Using Interval Skip Graphs Peter Desnoyers, Deepak Ganesan, and Prashant Shenoy Department of Computer Science.
An Integration Framework for Sensor Networks and Data Stream Management Systems.
A Metadata Based Approach For Supporting Subsetting Queries Over Parallel HDF5 Datasets Vignesh Santhanagopalan Graduate Student Department Of CSE.
EMP: A Network Management Protocol for IP-Based Wireless Sensor Networks 2010 International Conference on Communication in Wireless Environments and Ubiquitous.
U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Re-thinking Data Management for Storage-Centric Sensor Networks Deepak Ganesan University.
Action-Oriented Query Processing for Pervasive Computing Qiong Luo Joint work with Wenwei Xue Hong Kong University of Science and Technology (HKUST)
Mobile Relay Configuration in Data-Intensive Wireless Sensor Networks.
Turducken: Hierarchical Power Management for Mobile Devices Jacob Sorber, Nilanjan Banerjee, Mark Corner, Sami Rollins University of Massachusetts, Amherst.
EWatch: A Wearable Sensor and Notification Platform Paper By: Uwe Maurer, Anthony Rowe, Asim Smailagic, Daniel P. Siewiorek Presenter: Ke Gao.
Sensor Database System Sultan Alhazmi
1 EnviroStore: A Cooperative Storage System for Disconnected Operation in Sensor Networks Liqian Luo, Chengdu Huang, Tarek Abdelzaher John Stankovic INFOCOM.
Context-Awareness on Mobile Devices - the Hydrogen Approach Thomas Hofer, Wieland Schwinger, Mario Pichler, Gerhard Leonhartsberger, Josef Altmann (Software.
Query Processing for Sensor Networks Yong Yao and Johannes Gehrke (Presentation: Anne Denton March 8, 2003)
Department of Computer Science University of Massachusetts, Amherst TSAR*: A Two Tier Sensor Storage Architecture Using Interval Skip Graphs Peter Desnoyers,
International Directory Network (IDN) Scalability, Security and Interoperability WGISS, 2006 Tom Northcutt Systems Administrator: GCMD September 13, 2006.
IPower: An Energy Conservation System for Intelligent Buildings International Journal of Sensor Networks Yu-Chee Tseng, You-Chiun Wang, and Lun- Wu Yeh.
Enabling Large-Scale Storage in Sensor Networks with the Coffee File System ISPN 2009 Lawrence.
ELF: An Efficient Log-Structured Flash File System For Micro Sensor Nodes Hui Dai Michael Neufeld Richard Han University of Colorado at Boulder Computer.
Parallel Execution Plans Joe Chang
CLASS Information Management Presented at NOAATECH Conference 2006 Presented by Pat Schafer (CLASS-WV Development Lead)
Yanlei Diao, University of Massachusetts Amherst Future Directions in Sensor Data Management Yanlei Diao University of Massachusetts, Amherst.
Fuzzy Data Collection in Sensor Networks Lee Cranford Marguerite Doman July 27, 2006.
1 Compression and Storage Schemes in a Sensor Network with Spatial and Temporal Coding Techniques You-Chiun Wang, Yao-Yu Hsieh, and Yu-Chee Tseng IEEE.
Mining of Massive Datasets Ch4. Mining Data Streams
W. Hong & S. Madden – Implementation and Research Issues in Query Processing for Wireless Sensor Networks, ICDE 2004.
In-Network Query Processing on Heterogeneous Hardware Martin Lukac*†, Harkirat Singh*, Mark Yarvis*, Nithya Ramanathan*† *Intel.
University of Toronto Department of Electrical and Computer Engineering Jason Zebchuk and Andreas Moshovos June 2006.
Sensor Network and Smart Environment Research (SeNSER) Manohar Mareboyana Bo Yang.
An Energy-Efficient Approach for Real-Time Tracking of Moving Objects in Multi-Level Sensor Networks Vincent S. Tseng, Eric H. C. Lu, & Kawuu W. Lin Institute.
REED : Robust, Efficient Filtering and Event Detection in Sensor Network Daniel J. Abadi, Samuel Madden, Wolfgang Lindner Proceedings of the 31st VLDB.
Building Wireless Efficient Sensor Networks with Low-Level Naming J. Heihmann, F.Silva, C. Intanagonwiwat, R.Govindan, D. Estrin, D. Ganesan Presentation.
Efficient Opportunistic Sensing using Mobile Collaborative Platform MOSDEN.
Performing Fault-tolerant, Scalable Data Collection and Analysis James Jolly University of Wisconsin-Madison Visualization and Scientific Computing Dept.
Virtual Memory By CS147 Maheshpriya Venkata. Agenda Review Cache Memory Virtual Memory Paging Segmentation Configuration Of Virtual Memory Cache Memory.
STREAMS & SENSOR NETWORKS “ Query Processing in Sensor Networks ”
Mining High-Speed Data Streams Presented by: William Kniffin Pedro Domingos Geoff Hulten Sixth ACM SIGKDD International Conference
Software Architecture of Sensors. Hardware - Sensor Nodes Sensing: sensor --a transducer that converts a physical, chemical, or biological parameter into.
SOURCE:2014 IEEE 17TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND ENGINEERING AUTHER: MINGLIU LIU, DESHI LI, HAILI MAO SPEAKER: JIAN-MING HONG.
- Pritam Kumat - TE(2) 1.  Introduction  Architecture  Routing Techniques  Node Components  Hardware Specification  Application 2.
KISS-Tree: Smart Latch-Free In-Memory Indexing on Modern Architectures
Outline Ming Li, Deepak Ganesan and Prashant Shenoy, PRESTO: Feedback-Driven Data Management in Sensor Networks, Proceedings of the Third ACM/USENIX.
Outline Ganesan, D., Greenstein, B., Estrin, D., Heidemann, J., and Govindan, R. Multiresolution storage and search in sensor networks. Trans. Storage.
Presentation transcript:

1 Rethinking Data Management for Storage-centric Sensor Networks Yanlei Diao, Deepak Ganesan, Gaurav Mathur, and Prashant Shenoy CIDR 2007 Proceedings of the Third Biennial Conference on Innovative Data Systems Research (CIDR), Asilomar, CA, January 2007.

2 STONES STONES Project STONES  STO rage for N etworked E mbedded S ystems  Energy-efficient Storage for Sensors  Sensor Database  PRESTO, TSAR, Capsule, …etc

3 Papers PRESTO: Feedback-Driven Data Management in Sensor Networks ACM/USENIX NSDI 2006 TSAR: A Two Tier Sensor Storage Architecture Using Interval Skip Graphs ACM Sensys 2005 Capsule: An Energy-Optimized Object Storage System for Memory-Constrained Sensor Devices ACM Sensys 2006 Proxy Cache Sensor Nodes

4 PRESTO Introduction PRESTO Proxy PRESTO Sensor precision(query) > confidence interval x(t+1)-x(t) > worst-case deviation

5 TSAR Introduction

6 Capsule Introduction

7 Outline Introduction StonesDB Architecture Local Database Distributed Data Management Current Status and Conclusions References

8 Introduction Data Management in Sensor Networks Live Data Management  Real-time queries  Only small window of data is important  Event detection and notification  Push-down Filters, AQP, …etc Archival Data Management  Database outside the sensor networks  View sensor networks as database  Analysis of past events, Historical trends

9 Introduction Example: Smart Home and Smart Biz Live Data Management? Archival Data Management?

10 Introduction Centralized Archival Data Management Internet DBMS Database Management System User Query Data Access Sensors with high data rate?! camera, acoustic, vibration… lossless aggregation… Low data rate, High query rate 22.1 ˚C 21.5 ˚C 21.8 ˚C ?

11 Introduction Storage-centric Archival Data Management Internet User Query Data Access Local Storage Flash Memory acoustic image Push query to sensors!  limited capabilities  flash memory efficiency High data rate, Low query rate ?

12 Introduction Sensor Node Hardware Today Mica2 mote  6 MHz Processor  4 KB RAM  128 KB FLASH iMote2  13 – 416 MHz Processor  32 MB RAM  32 MB FLASH

13 Introduction Technology Trends Communication Storage Energy cost of storage compared to that of communication Generation of Sensor Platforms Energy Cost (per byte)

14 Design Goals Exploit local flash memory  Cheap, energy-efficient flash memory Exploit resources-rich proxies  cache data and split query plans Support a rich set of queries  SQL-type queries  data mining and similarity search queries Support heterogeneity  configurable to heterogeneous sensor platforms

15 StonesDB Architecture Two-tier Sensor Networks Local Database Distributed Data Management Layer user specified confidence bound

16 StonesDB Architecture System Operations Image Retrieval … Proxy Cache of Image Summaries 找出沒有洋蔥頭臉部表 情的圖片...

17 StonesDB Architecture System Operations Image Retrieval 找出沒有洋蔥頭臉部表 情的圖片... Query Engine Partitioned Access Methods … Sensor Local Storage

18 Local Database Architecture of Local Database Stream Index Summary

19 Local Database Costs and Benefits of Access Methods Cost for B+ tree insertion  Cost for sequential scan  H-level B+ tree cost for page read/write readings per page Sequential scan is 340 times more energy efficiency! When depth of B+ tree is 2… Scan is better when data is not accessed very frequently. Lazy index construction!

20 Local Database Partitioned Access Methods temporal segments B+ TreeR Tree Write-Once Indexing!

21 Local Database Summarization and Aging All available storage gets filled…  When to drop these summaries?  How to drop these summaries?  Graceful query quality degradation. local storage capacity Resolution 4 Resolution 1Resolution 2 Resolution 3 Multi-resolution Summarization: Local Storage Allocation

22 Local Database Summarization and Aging High query accuracy Low compactness Low query accuracy High compactness How long should a summary be stored in the network?

23 Local Database Summarization and Aging Query Accuracy Time Quality Difference present past 95% 50% user-desired quality degradation system-provided step function Objective: minimize the worst case quality difference

24 Distributed Data Management The Problems Proxy Cache of Image Summaries What summaries to cache? What resolution of summaries? How should a query plan be split? I want the data of …

25 Distributed Data Management Querying the Proxy Cache Internet User Query Statistical models? Low-resolution data? Metadata of images? Response Gateway, Proxy Sensor Nodes summaries from sensors to the proxy queries from the proxy to sensors query execution at the sensors results back to the proxy min

26 Distributed Data Management Querying the Sensor Tier Gateway, Proxy Sensor Nodes Cache miss… Not meet accuracy requirement? User Query How to split the query plan? Use Query Processing Engine…

27 Distributed Data Management Querying the Sensor Tier Sensor Nodes Gateway, Proxy Number of cars over past half hour? Proxy Cache of Image Summaries Partially process the query at the proxy!

28 Distributed Data Management Querying the Sensor Tier Sensor Nodes Gateway, Proxy Average temperature between PM 1:00 - 2: Proxy Cache of Data Summaries Refine the result at the sensor node! … 19.2 ˚C 19.5 ˚C 20.2 ˚C 21.1 ˚C 20.0 ˚C 18.6 ˚C average temperature every two hours

29 Current Status and Conclusions Implemented Capsule  Flash-based object store  Energy-efficient data structure (lists, arrays, trees)  Currently extended with summarization, aging, and partitioned indexing. Other related systems  TSAR: separate data from metadata  PRESTO: implement a proxy cache No running system for StonesDB architecture.

30 References The STONES Project STONES  CAPSULE  PRESTO  The Wireless Sensor Networks Group at UMASS 